From: "Andres Lagar-Cavilla" <andres@lagarcavilla.org>
To: Ian Campbell <Ian.Campbell@citrix.com>
Cc: George Dunlap <george.dunlap@eu.citrix.com>,
Olaf Hering <olaf@aepfle.de>,
"xen-devel@lists.xensource.com" <xen-devel@lists.xensource.com>
Subject: Re: [PATCH] RFC: initial libxl support for xenpaging
Date: Mon, 27 Feb 2012 06:45:38 -0800 [thread overview]
Message-ID: <9ca33a90d00fa202b73ba94abe14a2b5.squirrel@webmail.lagarcavilla.org> (raw)
In-Reply-To: <1330335877.8557.253.camel@zakaz.uk.xensource.com>
> On Fri, 2012-02-24 at 17:12 +0000, Andres Lagar-Cavilla wrote:
>> Sharing + balloon is mostly a bad idea. It's not forbidden or broken
>> from the hypervisor angle, but it's a lose-lose game. Ballooning a
>> shared page gains you nothing. The ballooner can't know what is shared
>> and what isn't, in order to make an educated decision. And the sharer
>> can't foresee what the balloon will victimize.
>
> Doesn't ballooning generally evict unused pages, because it allocates
> new _free_ pages, even if they were shared is there any benefit to doing
> so?
Certainly the balloon will pick free pages. The sharing daemon should not
have shared those, but it's not unlikely that it will have.
It's a classic semantic gap problem, and we were discussing this in the
context of paging ("the pager should not have paged out page table pages")
Beyond free pages, a prime target for sharing are read-only disk buffers
in the page cache. Those are victim #2 for the balloon.
>
> Or is it more of a second order effect, e.g. ballooning can cause the
> guest to swap out stuff which could have been shared?
>
Absolutely.
Andres
> Ian.
>
>
next prev parent reply other threads:[~2012-02-27 14:45 UTC|newest]
Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-02-16 7:47 [PATCH] RFC: initial libxl support for xenpaging Olaf Hering
2012-02-17 13:44 ` Ian Campbell
2012-02-17 14:25 ` Olaf Hering
2012-02-17 14:58 ` Ian Campbell
2012-02-17 15:24 ` Olaf Hering
2012-02-17 15:33 ` Ian Campbell
2012-02-17 15:43 ` Olaf Hering
2012-02-17 15:54 ` Ian Campbell
2012-02-17 16:03 ` Olaf Hering
2012-02-17 16:43 ` Ian Campbell
2012-02-20 10:44 ` George Dunlap
2012-02-20 11:12 ` Olaf Hering
2012-02-20 14:48 ` George Dunlap
2012-02-20 15:19 ` Olaf Hering
2012-02-20 16:06 ` George Dunlap
2012-02-20 15:38 ` Olaf Hering
2012-02-20 16:04 ` George Dunlap
2012-02-21 9:59 ` Ian Campbell
2012-02-21 11:27 ` Olaf Hering
2012-02-21 11:34 ` Ian Campbell
2012-02-21 12:21 ` Olaf Hering
2012-02-21 12:20 ` George Dunlap
2012-02-23 10:42 ` Ian Campbell
2012-02-23 12:18 ` George Dunlap
2012-02-23 16:22 ` Andres Lagar-Cavilla
2012-02-23 17:30 ` George Dunlap
2012-02-24 10:11 ` Ian Campbell
2012-02-24 15:38 ` Olaf Hering
2012-02-24 16:39 ` Ian Campbell
2012-02-24 22:19 ` Olaf Hering
2012-02-24 17:12 ` Andres Lagar-Cavilla
2012-02-27 9:44 ` Ian Campbell
2012-02-27 14:45 ` Andres Lagar-Cavilla [this message]
2012-02-28 13:05 ` George Dunlap
2012-02-28 15:25 ` Andres Lagar-Cavilla
[not found] ` <CAFLBxZapKUVaiz4e-VwGTwj0a0FOPij5hXrRkevrTC6brtug4Q@mail.gmail.com>
2012-02-28 17:03 ` Fwd: " George Dunlap
2012-03-06 23:07 ` Andres Lagar-Cavilla
[not found] ` <CAFLBxZavBR0c_p1E9NpfkNzhrN-LCbtV=kPRwFZ3VFFMKPBNKg@mail.gmail.com>
2012-02-28 17:02 ` Fwd: " George Dunlap
2012-02-28 17:12 ` Ian Campbell
2012-02-21 9:05 ` Ian Campbell
[not found] <mailman.3929.1329497068.1471.xen-devel@lists.xensource.com>
2012-02-17 16:55 ` Andres Lagar-Cavilla
2012-02-17 17:03 ` Ian Campbell
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=9ca33a90d00fa202b73ba94abe14a2b5.squirrel@webmail.lagarcavilla.org \
--to=andres@lagarcavilla.org \
--cc=Ian.Campbell@citrix.com \
--cc=george.dunlap@eu.citrix.com \
--cc=olaf@aepfle.de \
--cc=xen-devel@lists.xensource.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).